User Sentiment Research Articles

Analysis Sentiment of Twitter User on Indonesia's 2024 Presidential Election Using K-Means Algorithm

The General Election is a five-year agenda of the Indonesian people in order to fulfill the political rights of every citizen for the election of the president and the legislature. In every election, especially during the campaign period, differences of opinion often occur between certain groups or factions, this often creates an atmosphere of political uproar in various parts of Indonesia. The purpose of this study is to see the level of sentiment of social media users towards the implementation of elections in Indonesia so as to minimize political upheaval that occurs in society during the elections to be held in 2024. The data that will be used in this research is data on Twitter users who have large volumes and are taken from all regions of Indonesia. To suit the data model used, this study will use the data mining method with the K-Means algorithm. The results of this study show the percentage level of public sentiment of Twitter users towards the 2024 election and presidential election. Public sentiment is positive, neutral and negative. Based on these results, it can provide input to the government so that it can make appropriate policies ahead of elections and presidential elections so as to create a peaceful atmosphere.

Sentiment Analysis and Topic Modeling of Kitabisa Applications using Support Vector Machine (SVM) and Smote-Tomek Links Methods

Kitabisa is an Indonesian application that functions to raise funds online. Users can easily support various types of campaigns and donate funds to various social causes through the app. User reviews of the application are very diverse, and it is not sure whether user reviews of the application tend to be positive, neutral, or negative. This research aimed to analyze the sentiment of the Kitabisa application by modeling topics using Latent Dirichlet Allocation (LDA) and classifying user reviews using a Support Vector Machine (SVM). The scrapped dataset showed imbalanced dataset problems, so the SMOTE-Tomek Links oversampling technique was proposed. The results of this study show that using LDA produces five topics often discussed in 750 reviews. Then, the performance of SVM without using SMOTE-Tomek Links was 72% accuracy, 76% precision, 72% recall, and 64% f1 score. Meanwhile, using SMOTE-Tomek Links could significantly improve the performance, namely 98% accuracy, 98% precision, 98% recall, and 98% f1 score. Based on this research, the application of SVM achieved high performance for user sentiment classification, especially when the dataset was in a balanced state. Therefore, the SMOTE-Tomek Links oversampling technique is recommended for dealing with unbalanced sentiment datasets.

Content analysis of YouTube videos regarding natural disasters in India and analysis of users sentiment through viewer comments

Content analysis of YouTube videos regarding natural disasters in India and analysis of users sentiment through viewer comments

Unpacking the Drivers of Dissatisfaction and Satisfaction in a Fitness Mobile Application.

This research investigates the factors influencing user satisfaction and dissatisfaction in fitness mobile applications. It employs Herzberg's two-factor model through text mining to classify Fitbit mobile app attributes into satisfiers and dissatisfiers. The Fitbit app was chosen due to its prevalence in the United States. The study analyzes 100,000 English reviews from the Fitbit app on the Google Play Store, categorizing attributes. It identifies three dissatisfying categories (functional, compatibility, paid services) and three satisfying categories (gratification, self-monitoring, self-regulation), comprising 25 sub-attributes. This classification offers in-depth insights into what drives user contentment or discontent with fitness apps. The findings contribute to the fitness app domain by applying text-mining and Herzberg's model. Researchers can build upon this foundation, and practitioners can use it to enhance app experiences. However, this research relies on user reviews, often lacking comprehensive explanations. This limitation may hinder a profound understanding of the underlying psychological aspects in user sentiments. Nonetheless, this study takes strides toward optimizing fitness apps for users and developers.

Sentiment Analysis of Honkai: Star Rail Indonesian Language Reviews on Google Play Store Using Bidirectional Encoder Representations from Transformers Method

Online games are a type of entertainment that is done by humans to have fun and forget all the problems in everyday life. Honkai: Star Rail is a new online game application owned by miHoYo which is currently popular and widely downloaded on the Google Play Store. Reviews on the Honkai: Star Rail app are increasing over time so this makes it difficult for app developers to know past user reviews on their apps. Therefore, the author conducted a study to analyze sentiment towards Honkai: Star Rail application reviews in Indonesian on the Google Play Store using the Bidirectional Encoder Representations from Transformers (BERT) method to determine user sentiment towards the Honkai: Star Rail application and then processed further so that it becomes a record for developers, users, and prospective users of the Honkai: Star Rail application. This study uses Indonesian language review data from users of the Honkai: Star Rail application found on the Google Play Store website as many as 6000 reviews. The BERT method applied in this study consisted of data collection, dataset labeling, data preprocessing, dataset splitting, modeling, model training, and evaluation. Based on the evaluation results that have been carried out on the test data, 97 data are true positive with 27 data are false positive, 4 data are true neutral with 47 data are false neutral, and 381 data are true negative with 37 data are false negative. So, it can be concluded that the model still has difficulty predicting reviews with neutral sentiment but is good enough at predicting reviews with positive and negative sentiment. In addition, the accuracy of the model is 81% with a precision of 63% for positive sentiment reviews, 36% for neutral sentiment reviews, and 89% for negative sentiment reviews.

Users Sentiment Analysis Using Artificial Intelligence-Based FinTech Data Fusion in Financial Organizations

Users Sentiment Analysis Using Artificial Intelligence-Based FinTech Data Fusion in Financial Organizations

ASPECT-BASED SENTIMENT ANALYSIS TERHADAP ULASAN APLIKASI FLIP MENGGUNAKAN PEMBOBOTAN TERM FREQUENCY-INVERSE DOCUMENT FREQUENCY (TF-IDF) DENGAN METODE KLASIFIKASI K-NEAREST NEIGHBORS (K-NN)

The rapid growth of online transactions in Indonesia has increased the demand for efficient interbank transfer solutions. However, the costs associated with such transactions have become a significant obstacle. Flip, a company with a vision to become a global leader in customer satisfaction-driven services, offers a solution to this challenge. This study proposes an aspect-based sentiment analysis method using the K-Nearest Neighbors (K-NN) algorithm to analyze user sentiment on key aspects, namely speed, security, and the cost of using the Flip application. The results of this research provide valuable information that can be used as a basis to provide insights, suggestions and recommendations to businesses, so they can create better solutions and promote optimal user experience. The research results show that the K-NN model has the ability to predict user psychology well in all aspects, with a significant level of accuracy, specifically speed (73.04%), security (86, 05%) and costs (80.11%). In addition, this study also compares two model validation methods: simple data splitting method and K-Fold cross-validation. Although the simple data splitting method has a higher average accuracy, K-fold cross-validation is considered superior as it provides a more accurate and reliable estimate of the overall performance of the model. Sentiment analysis results show that Flip app users tend to give negative feedback on speed and security, while they give positive feedback on cost. Therefore, the main recommendation is that the company PT Fliptech Lentera Inspirasi Pertiwi improves the speed and security aspects to increase user satisfaction with the Flip application. Therefore, this customer-centric service will continue to prioritize user satisfaction as its primary goal.

Characterizing Twitter chatter about temporary alcohol abstinence during "Dry January".

With roots as a public health campaign in the United Kingdom, "Dry January" is a temporary alcohol abstinence initiative encouraging participants to abstain from alcohol use during the month of January. Dry January has become a cultural phenomenon, gaining increasing news media attention and social media engagement. Given the utility of capturing naturalistic discussions around health topics on social media, we examined Twitter chatter about Dry January and associated temporary abstinence experiences. Public tweets were collected containing the search terms "dry january" or "dryjanuary" posted between 15 December and 15 February across 3years (2020-2). A random subsample stratified by year (n = 3145) was pulled for manual content analysis by trained coders. Final codebook accounted for user sentiment toward Dry January, user account type, and themes related to Dry January participation. Engagement metadata (e.g. likes) were also collected. Though user sentiment was mixed, most tweets expressed positive or neutral sentiment toward Dry January (74.7%). Common themes included encouragement and support for Dry January participation (14.1%), experimentation with and promotion of nonalcoholic drinks (14.0%), and benefits derived from Dry January participation (10.4%). While there is promise in the movement to promote positive alcohol-related behavior change, increased efforts to deliver the campaign within a public health context are needed. Health communication campaigns designed to inform participants about evidence-based treatment and recovery support services proven to help people quit or cut down on their drinking are likely to maximize benefits.

From YouTube Comments to Insights: A Sentiment Analysis of Opinions on Productivity Tools

Abstract: YouTube, a diverse content hub, acts as a prominent platform for both viewing videos and engaging through comments that express opinions. By carefully analyzing these comments, sentiments associated with products or topics become evident. This research delves into sentiment analysis of YouTube comments, specifically focusing on sentiments towards productivity tools and technologies. Through this examination of conveyed sentiments, it deepens understanding of public perceptions and attitudes toward these tools. Utilizing YouTube's reach and engagement, the study systematically collects comment data using the YouTube API and employs the established VADER sentiment analysis framework. Employing a Bag-of-Words approach, the research incorporates machine learning algorithms like Naive Bayes and Random Forest, achieving accuracy rates of 84.01% and 91.34%, respectively. In-depth temporal analysis of user engagement patterns uncovers trends, with heightened engagement followed by decline, correlating with external events, notably the Covid-19 pandemic. These insights enhance sentiment analysis and illuminate dynamics between societal occurrences, user sentiments, and digital dialogues, offering perspectives on evolving opinions about productivity tools and technologies

SENTIMENT ANALYSIS OF MYPERTAMINA APPLICATION USING SUPPORT VECTOR MACHINE AND NAÏVE BAYES ALGORITHMS

In line with the needs of the community and the progress of the times in the advanced field of fintech, cash payments are currently considered insecure as well as ineffective and efficient. To run a non-cash or cashless transaction program currently run by the government, PT. Pertamina invites the public to use E-Payment from the My Pertamina application in collaboration with LinkAja. In this study, the sentiments of MyPertamina application users will be analyzed based on reviews on the Google Play Store. Review data will be analyzed to determine whether the review has positive, negative, or neutral sentiments. The data analysis stage is text preprocessing to change uppercase to lowercase, clearing text, separating text, taking important words, changing essential words, and labeling data into positive, negative, and neutral classes. As well as the classification and evaluation of results. This study used the Support Vector Machine (SVM) and Naïve Bayes classification methods. To evaluate the results, the confusion matrix was used to test the accuracy, precision, recall, and F1 score value. The classification results obtained the highest accuracy value for the Support Vector Machine (SVM) method, which had accuracy (68.50%), precision (70.00%), recall (69.70%), and F1 score (68.46%). Meanwhile, the Naïve Bayes method has performance with accuracy (63.00%), precision (63.90%), recall (61.34%), and F1 score (59.55%).

SENTIMENT ANALYSIS OF SERVICE PROVIDER ON TWITTER TWEET USING NAIVE BAYES CLASSIFIER WITH PHP

This study aimed to analyze the sentiments of Twitter users towards internet services provided by Indosat, Ooredoo and XL Axiata in Indonesia. Twitter Tweet data is taken using the Twitter API and processed using the PHP programming language. The classification process uses the Naïve Bayes Classifier algorithm to group positive, negative, and neutral categories. The calculation results show that negative sentiment towards Indosat Ooredoo services scores 56%, while sentiment towards XL Axiata services negatively scores 50%. The accuracy of the sentiment analysis system using PHP reaches 78% based on comparison with manual classification results. This research provides benefits in the form of information about user sentiment towards cellular internet service providers, so that each provider can improve and optimize internet services based on this data. 
 Keywords: twitter, sentiment analysis, naïve bayes, php

Textual emotion analysis using improved metaheuristics with deep learning model for intelligent systems

AbstractComplex, intelligent systems, namely social network services (SNS), internet‐of‐things (IoT), augmented reality (AR), virtual reality (VR), and so on have become a crucial part of our day‐to‐day life. The dynamics of this system alter human behaviors in various forms. Therefore, it is important to understand the fundamental dynamics of the system and what it involves for user sentiment. Emotion analysis is a commonly used and very powerful analysis methodology in data mining. It gives an outstanding option to evaluate, determine, understand, and monitor the sentiments of consumers with respect to a service or a product. This article focuses on the design of textual emotion analysis using the improved metaheuristics with deep learning (TEA‐IMWDL) model for intellectual systems. The goal of the TEA‐IMWDL technique lies in the identification and classification of emotions that exist in social media content. To attain this, the presented TEA‐IMWDL technique undergoes preprocessing to make the input data compatible with the latter stages of processing. For the word embedding process, the Fast text approach is utilized in this study. Next, the long short term memory‐recurrent neural network (LSTM‐RNN) method can be enforced for the emotion classification procedure. To boost the classification efficiency of the LSTM‐RNN model, an improved gravitational search optimization algorithm (IGSA) based hyperparameter tuning process is involved in this study. A brief set of simulation analyses of the TEA‐IMWDL technique is carried out and the results are tested on a benchmark dataset. The experimental outcomes demonstrate the enhanced performance of the TEA‐IMWDL technique over other recent approaches.

Leveraging Sentiment Analysis for Twitter Data to Uncover User Opinions and Emotions

Huge amounts of emotion are expressed on social media in the form of tweets, blogs, and updates to posts, statuses, etc. Twitter, one of the most well-known microblogging platforms, is used in this essay. Twitter is a social networking site that enables users to post status updates and other brief messages with a maximum character count of 280. Twitter sentiment analysis is the application of sentiment analysis to Twitter data (tweets) in order to derive user sentiments and opinions. Due to the extensive usage, we intend to reflect the mood of the general people by examining the thoughts conveyed in the tweets. Numerous applications require the analysis of public opinion, including businesses attempting to gauge the market response to their products, the prediction of political outcomes, and the analysis of socioeconomic phenomena like stock exchange. Sentiment classification attempts to estimate the sentiment polarity of user updates automatically. So, in order to categorize a tweet as good or negative, we need a model that can accurately discern sarcasm from the lexical meaning of the text. The main objective is to create a practical classifier that can accurately classify the sentiment of twitter streams relating to GST and Tax. Python is used to carry out the suggested algorithm.

Comparing the Performance of Data Mining Algorithms in Predicting Sentiments on Twitter

On Twitter, users can post tweets, videos, and images. It can, however, also be disruptive and difficult. To categorize the material and improve searchability, hashtags are crucial. This study focuses on examining the opinions of Twitter users who participate in trending topics. The algorithms K-Nearest Neighbor (KNN) and Support Vector Machine (SVM) are used for sentiment analysis. The data set comprises tweet information on popular topics that was collected using the Twitter API and saved in Excel format. SVM and K-NN are used for data preparation, weighting, and sentiment analysis. With 105 data points, the study provides insight into user sentiment. SVM identified 99% of positive responses and 1% of negative responses with an accuracy of 80%. KNN successfully identified 90% of the positive responses and 10% of the negative responses, with an accuracy rate of 71.4%. According to the results, SVM performs better when analyzing the sentiment of hashtag users on Twitter.

Sentimental Analysis Of COVID-19 Vaccination Tweets Using Machine Learning Techniques

Twitter Sentiment Analysis applies sentiment analysis to data in tweets on social media platforms to identify user sentiment. The scope of research in this field has expanded steadily in recent decades. The reason is that the format of tweets is complex and difficult to process. The extremely small Tweet format introduces a whole new set of problems: B. Use of Slang and Acronyms. The global community remains affected by the ongoing COVID-19 pandemic, with its profound implications for the health and overall welfare of individuals worldwide. Our worldview has changed as a result of the pandemic. A vaccination campaign should be carried out among the population to stop the outbreak of the pandemic. However, people are unsure about vaccination, which is cause for concern. In this study, machine learning techniques were employed to assess the sentiment of public Twitter tweets concerning COVID- 19 vaccination. The researchers utilized two distinct ML methods, namely Support Vector Machine (SVM) and Logistic Regression (LR), to analyze the Twitter data and classify the tweets as either positive or negative. Keywords:Classifier, Twitter, Sentiment Analysis.

SENTIMENT CLASSIFICATION OF TWEETS WITH EXPLICIT WORD NEGATIONS AND EMOJI USING DEEP LEARNING

The widespread use of social media platforms such as Twitter, Instagram, Facebook, and LinkedIn have had a huge impact on daily human interactions and decision-making. Owing to Twitter's widespread acceptance, users can express their opinions/sentiments on nearly any issue, ranging from public opinion, a product/service, to even a specific group of people. Sharing these opinions/sentiments results in a massive production of user content known as tweets, which can be assessed to generate new knowledge. Corporate insights, government policy formation, decision-making, and brand identity monitoring all benefit from analyzing the opinions/sentiments expressed in these tweets. Even though several techniques have been created to analyze user sentiments from tweets, social media engagements include negation words and emoji elements that, if not properly pre-processed, would result in misclassification. The majority of available pre-processing techniques rely on clean data and machine learning algorithms to annotate sentiment in unlabeled texts. In this study, we propose a text pre-processing approach that takes into consideration negation words and emoji characteristics in text data by translating these features into single contextual words in tweets to minimize context loss. The proposed preprocessor was evaluated on benchmark Twitter datasets using four deep learning algorithms: Long Short-Term Memory (LSTM), Recurrent Neural Network (RNN), and Artificial Neural Network (ANN). The results showed that LSTM performed better than the approaches already discussed in the literature, with an accuracy of 96.36%, 88.41%, and 95.39%. The findings also suggest that pre-processing information like emoji and explicit word negations aids in the preservation of sentimental information. This appears to be the first study to classify sentiments in tweets while accounting for both explicit word negation conversion and emoji translation.

Analisis Sentimen Pengguna Twitter Terhadap Grup Musik BTS Menggunakan Algoritma Support Vector Machine

BTS is a music group from South Korea which has global popularity around the world. As a music group that has global popularity, BTS influence on social media is quite large, with this influence making many companies interested in making them Brand Ambassadors (BA). A company can do a sentiment analysis of the public figure they want to make BA, later this sentiment can be taken into consideration for them to be able to determine the BA of their product. Sentiment analysis can also be used to distinguish the attitude of customers, users or followers towards a brand, topic, or product with the help of their reviews. Based on this, this study will analyze the sentiments of Twitter users towards music group BTS, using the Knowledge Discovery Database (KDD) research methodology, with 5 stages namely Data Selection, Data Preprocessing, Data Transformation, Text Mining and Evaluation. By using the Support Vector Machine (SVM) algorithm with a linear kernel, this study will do 3 scenarios with the distribution of training data and testing data 90:10 in scenario 1, 80:20 in scenario 2, and 70:30 in scenario 3. Confusion Matrix is used to evaluate the performance of the algorithm used and the results show that the best performance of the model formed is in scenario 1 and scenario 2.

Sentiment analysis on Twitter data towards climate action

Understanding the progress of the Sustainable Development Goals (SDGs) proposed by the United Nations (UN) is important, but difficult. In particular, policymakers would need to understand the sentiment within the public regarding challenges associated with climate change. With this in mind and the rise of social media, this work focuses on the task of uncovering the sentiment of Twitter users concerning climate-related issues. This is done by applying modern natural-language-processing (NLP) methods, i.e. VADER, TextBlob, and BERT, to estimate the sentiment of a gathered dataset based on climate-change keywords. A transfer-learning-based model applied to a pre-trained BERT model for embedding and tokenizing with logistic regression for sentiment classification outperformed the rule-based methods VADER and TextBlob; based on our analysis, the proposed approach led to the highest accuracy: 69%. The collected data contained significant noise, especially from the keyword ‘energy’. Consequently, using more specific keywords would improve the results. The use of other methods, like BERTweet, would also increase the accuracy of the model. The overall sentiment in the analyzed data was positive. The distribution of the positive, neutral, and negative sentiments was very similar in the different SDGs.

Effective Information Retrieval Framework for Twitter Data Analytics

The widespread adoption of opinion mining and sentiment analysis in higher cognitive processes encourages the need for real time processing of social media data to capture the insights about user's sentiment polarity, user's opinions, and current trends. In recent years, lots of studies were conducted around the processing of data to achieve higher accuracy. But reducing the time of processing still remained challenging. Later, big data technologies came into existence to solve these challenges but those have its own set of complexities along with having hardware deadweight on the system. The contribution of this article is to touch upon mentioned challenges by presenting a climbable, quick and fault tolerant framework to process real-time data to extract hidden insights. This framework is versatile enough to support batch processing along with real time data streams in parallel and distributed environment. Experimental analysis of proposed framework on twitter posts concludes it as quicker, robust, fault tolerant, and comparatively more accurate with traditional approaches.

A heterogeneous network embedded medicine recommendation system based on LSTM

In the healthcare sector, patient data plays a crucial role in medical diagnoses and treatment plans. However, existing techniques for finding similar patients based on Electronic Health Record (EHR) data face challenges due to high-dimensional and sparse vectors. To overcome this challenge, the paper proposes developing a novel heterogeneous network-embedded drug recommendation system. The system focuses on classifying the sentiment of drug users based on their reviews and other relevant features such as their medical condition, drug rating, and usage date. The overall framework of the system follows a step-by-step approach, starting with data exploration and preprocessing, followed by the development of a classification model based on Long-Short-Term Memory (LSTM) networks. During the data exploration phase, various visualization and statistical techniques are employed to analyze the different data types. This process helps in understanding the characteristics of the data, identifying patterns, and preparing the data to align with the research objective. Furthermore, additional variables are considered suitable for the LSTM model, a recurrent neural network (RNN) type designed to handle sequence data and long-term prediction problems. Unlike other models that process individual data points, LSTM incorporates feedback connections to process complete data sequences. This approach enhances the effectiveness of recommendation systems and enables the prediction of new drug user ratings based on existing user ratings. The developed system demonstrates promising results, achieving a classification accuracy of 92%. This indicates its ability to accurately predict the sentiment of drug users based on their reviews and other associated features. The novelty of this research lies in the integration of a heterogeneous network-embedded approach with LSTM-based classification, providing a more comprehensive and accurate drug recommendation system compared to existing methods.